Mobile radiocommunications

Outline :
1. Mobile (receiver) sensitivity
2. General propagation model
2.1 Shadowing
2.2 Fading
3. Empirical model
4. Exercices

Exercise session 12 : Mobile radiocommunications

1. Mobile (receiver) sensitivity

The transmitter is characterized by its power.

The receiver is characterized by its sensitivity.
After demodulation :

C EbW
Eb
=
=
N N0W N0
Sensitivity = minimum value of C such that
Eb
>
N0

Eb
N0


threshold

Therefore,

S=

Eb
N0


+N
threshold

For the mobile device,

W = 271 kHz N = kT0W = 120 dBm
for T0 = 290K , to which we add 10 dB (noise generated by the input
amplifier).
In addition,


Eb
N0


= 8 dB

Exercise session 12 : Mobile radiocommunications

It comes
S = 8 120 + 10 = 102 dBm

For a base station, we use generally

S = 104 dBm

Two communicating entities (base station, mobile device) may

demonstrate different transmission power and sensitivity.
Summary table (typical values) :
Receiver type

Sensitivity [dBm]

Base station
Mobile 8W
Mobile 2W
Two-band mobile

-104
-104
-102
-102

Exercise session 12 : Mobile radiocommunications

2. General propagation model

Generally speaking,
PR = PT LT + GT L + GR LR
where
LT = losses in the transmission circuits.
LR = losses in the receiving circuits.
L = free space losses.
In the case of an unique direct path :

L=

4d

2

Friis equation. In this case,

L = 32, 5 + 20 log f [MHz] + 20 log d [km]
But in practical situations :
Fading : Multipath, due to reflections and diffractions by
present objects.
Doppler effect, due to the mobile device movements.
Shadowing : attenuation due to present objects.
The transmission channel is constantly changing due to the
mobile movement.
Exercise session 12 : Mobile radiocommunications

2.1 Shadowing

Due to obstruction by objects.

Also called slow fading.
If we assume the presence of N obstacles. Then
Lshadowing[dB] = L1 + L2 + . . . + LN
with Li (i = 1, 2, . . . , N) = RV of same caracteristics. Then L follow a
normal law (central limit theorem) :
Lshadowing [dB] = N L50%, s2

where L50% is the median value of the attenuation (given by the

empirical models coming from the experimental data).

Exercise session 12 : Mobile radiocommunications

Impact on the covering zone (cell border) :

Lshadowing [dB] = L50% [dB] + Ls [dB]
In practice, we add a security margin Ms to the transmitted power.
The covering probability is given by p(Ls < Ms) and is then a function
of Ms .
Covering probability (in percent)
100

90

80

70

60

50

40

30

20

10

0
0

10

12

14

Margin in [dB]

Exercise session 12 : Mobile radiocommunications

2.2 Fading

Due to multipathes :

A cos(2 fot)

x(t)

Generally, there is no direct path.

Received signal :
X(t) = Ci cos(2 f0t + i)
i

which may be written in the following form

X(t) = XI (t) cos(2 f0t) XQ(t) sin(2 f0t)
with
XI (t) = Ci cos i et XQ(t) = Ci sin i
i

XI (t) and XQ(t) = gaussian and centered RV.

Exercise session 12 : Mobile radiocommunications

Therefore,
X(t) = R(t) cos (2 f0t + (t))
with
R(t) =

XI (t) + XQ(t)

and
(t) = tan1

XQ(t)
, (t) [0, 2[
XI (t)

Probability density of R(t) :

fR(t)(r) =

r
e
X2

2
r2
2

if r 0
if r < 0

0.7
0.6
0.5
0.4
0.3
0.2
0.1
0

Exercise session 12 : Mobile radiocommunications

Special caracteristics :
E {R(t)} = 1, 253 X
2
R(t)
= 0, 429 X

Probability density of (t) :


f(t)( ) =

1
2

Exercise session 12 : Mobile radiocommunications

if [0, 2[
otherwise

3. Empirical models

provide reliable order of magnitude for a lot of reference configuration.

Environnement types
Models example :

Rural
Suburban
Urban

BS

L at 925 [MHz]

L at 1795 [MHz]

100
100
50

90, 9 + 31, 8 log d

95, 9 + 31, 8 log d
123, 6 + 33, 8 log d

97, 0 + 31, 8 log d

102, 0 + 31, 8 log d
133, 1 + 33, 8 log d

where
BS = Base station antenna height (in [m] ).
d is expressed in kilometers.

Exercise session 12 : Mobile radiocommunications

10

Cell types
Macro cell : within a radius of a few dozen of [km] , in rural
environment.
Small cell : within a radius of a few [km] , urban environment.
Micro cell : within a radius of a few hundred [m] , dense urban
environment.
Pico cell : within a radius of a few dozen of [m] , indoor
environment.
Influence of the antenna height

hb

hm
11111111111111111111111111
00000000000000000000000000
00000000000000000000000000
11111111111111111111111111
d

L ' 20 log hb[m] + 20 log hm[m] + 40 log d[km]

Exercise session 12 : Mobile radiocommunications

11

Macro cells models

Formula of Cost 231-Hata :
L

46, 33 + 33, 9 log f [MHz]

13, 82 log hb[m] a(hm[m])
+(44, 9 6, 55 log hb[m]) log d[km] +Cm

with
a(hm) = (1, 1 log f 0, 7)hm (1, 56 log f 0, 8)
for a middle-sized tow, and

0 dB
Cm =
3 dB

middled-size towns
large cities

Formula correct in urban environment for cells of radius larger or

equal to 1 km and for frequency from 1500 to 2000 MHz .

Exercise session 12 : Mobile radiocommunications

12

Indoor propagation
Two types of propagation outdoor-indoor :
Soft Indoor : fading in places close to the front of the building,
typically 10 dB of additional fading.
Deep Indoor : fading in places located deep inside the buildings,
typically 20 dB of additional fading.

Exercise session 12 : Mobile radiocommunications

13

4. Exercices

1. A base station must communicate to a telecommunication mobile subscriber using

a mobile phone receiving in the 1800 [MHz] band, functionning at a nominal power
of 2 [W ] and comprising an isotropic antenna. The equivalent isotropically radiated
power of the base station is 50 [dBW ] .
(a) Considering that the communication take place in a middle-sized town, by using
the model Cost 231-Hata, compute the maximum radius of the cell covered
by the base station if its height is 40 [m] and by neglecting the effect due to the
height of the mobile.
(b) We would like to manage the shadowing effect. Determine the value of the
additionnal margin if we require a covering ratio of 90%.
(c) With this margin, and in case of Soft Indoor communications, what is the new
value of the maximum cell radius. We will also consider 3 [dB] losses due to
human bodies.
Remarks :
Following the Cost 231-Hata model, the fading Lu in urban environment is,
in [dB],
Lu = 46.33+33.9 log ( f )13.82 log (hb )a (hm )+[44.9 6.55 log (hb )] log (d)+Cm
with
f the frequency, d the distance, hb , hm the heights ; these values are
respectively expressed in [MHz], [km] and [m].
a (hm ) = (1.1 log ( f ) 0.7) hm (1.56 log ( f ) 0.8) for a middle-sized town.
Cm = 0 [dB] for middle-sized towns and suburbs, and Cm = 3 [dB] for large
cities

Exercise session 12 : Mobile radiocommunications

14

2. A GSM service provider would like to deploy a cellular network in a large city and
is interrested in the modelization of a circular cell. We would like to determine the
maximum radius of the cell, knowing that the transmission power of the base station
(BTS) is equal to 50 [W ] and that the used frequency is 1800 [MHz].
(a) Determine the maximum radius of the cell by using the Cost 231-Hata model
by neglecting the effects depending on the mobile height. The height of the base
station is 40 [m] .
(b) We would lile to guard ourselves
additional margin value if we require
to communicate in Soft Indoor and
to human bodies ? Determine again
conditions.

from shadowing effects. Determine the

a covering percentage of 90%, if we wish
if we also consider 3 [dB] of losses dues
the maximum radius of the cell in these

(c) In a second time, we focus our attention on the design in terms of the traffic of
the cell. We will arbitrarily choose a cell radius of 0.5 [km].
Knowing that :


the service provider covers 500 clients/km2 ,
10 % of the covered customers in the cell have establish a communication
during the observation period of 15 [min] and
the average call duration is 5 [min],
determine the simultaneous communication number that the base station have
to support if we suppose a blocking probability of 0.02.
(d) Determine the minimum spectral occupancy, knowing that each carrier is able to
transport a maximum of 8 calls.
Remarks :
The receiving and transmitting antenna are assumed isotropic
The margin values are assumed independent of the frequency
Following the Cost 231-Hata model, the fading Lu in urban environment is,
in [dB],
Lu = 46.33+33.9 log ( f )13.82 log (hb )a (hm )+[44.9 6.55 log (hb )] log (d)+Cm
with
f the frequency, d the distance, hb , hm the heights ; these values are
respectively expressed in [MHz], [km] and [m].
a (hm ) = (1.1 log ( f ) 0.7) hm (1.56 log ( f ) 0.8) for a middle-sized town.
Cm = 0 [dB] for middle-sized towns and suburbs, and Cm = 3 [dB] for large
cities

Exercise session 12 : Mobile radiocommunications

15

3. A mobile service provider analyzes, in a middle-sized town, the effect of the cell size
on the power received by the mobile devices.
(a) We assume that we work only in the 1800 [MHz] band and that the Cost
231-Hata model is valid. Compute the maximum radius of a cell.
Note : for the computation, we consider that the base height is 30 [m] and we
neglect the effects dues to the mobile height. The transmitting antenna has a
power of 100 [W ] and a transmission gain of 5 [dB]. We are interrested in a Deep
Indoor covering.
(b) If we double the radius of the (circular) cells ; analyze the effect of such
a modification on the transmission gain, when the other parameters stay
unchanged !
(c) To guard ourselves from some shadowing effects, the service provider decides
to multiply the EIRP by 3. What do you think of this solution ? What is the
covering percentage ?
Note : We consider the same conditions that in the point (b).
Remarks :
Following the Cost 231-Hata model, the fading Lu in urban environment is,
in [dB],
Lu = 46.33+33.9 log ( f )13.82 log (hb )a (hm )+[44.9 6.55 log (hb )] log (d)+Cm
with
f the frequency, d the distance, hb , hm the heights ; these values are
respectively expressed in [MHz], [km] and [m].
a (hm ) = (1.1 log ( f ) 0.7) hm (1.56 log ( f ) 0.8) for a middle-sized town.
Cm = 0 [dB] for middle-sized towns and suburbs, and Cm = 3 [dB] for large
cities

Exercise session 12 : Mobile radiocommunications

16

4. A GSM
 2  service provider would like to cover a middle-sized town on an area of
20 km with a number N of omnidirectional antennas having a power of 80 [W ], a
gain of 5 [dB] and a height of 40 [m].
This service provider enforces a Deep Indoor covering with 90% of covering percentage.
(a) What is the spectral efficiency of this GSM system ?
(b) Determine, at 900 [MHz], the minimum number N of antennas required to cover
the mentionned area if we suppose that these antennas cover the whole area
without holes nor covering. We assume that the losses dues to human bodies
are equal to 3 [dB]. We will use the Cost 231-Hata model by neglecting the
effects dependent on the mobile device height.
(c) Compare the result obtained above with the result when the frequency is equal
to 1800 [MHz]. Comment your answer.
(d) If the omnidirectional antennas are replaced by trisectoral antennas having the
same maximum gain than the omnidirectional antennas, could we place more
antennas ? What would be the advantages
Remarks :
The receiving antenna is supposed isotropic
The margin values are identical at 900 [MHz] and 1800 [MHz].
Following the Cost 231-Hata model, the fading Lu in urban environment is,
in [dB],
Lu = 46.33+33.9 log ( f )13.82 log (hb )a (hm )+[44.9 6.55 log (hb )] log (d)+Cm
with
f the frequency, d the distance, hb , hm the heights ; these values are
respectively expressed in [MHz], [km] and [m].
a (hm ) = (1.1 log ( f ) 0.7) hm (1.56 log ( f ) 0.8) for a middle-sized town ;
this correction factor depend on the mobile device antenna height but also
on the environment type.
Cm = 0 [dB] for middle-sized towns and suburbs, and Cm = 3 [dB] for the
large cities

Exercise session 12 : Mobile radiocommunications

17

Answer
1. (a) 23,9 km .
(b) 8 dB .
(c) 5,87 km .

Exercise session 12 : Mobile radiocommunications

18